When updating used ring, the id in used element should be the
index of the first desc in the desc chain.
Fixes: f9b9d1a557 ("net/virtio-user: add multiple queues in device emulation")
Cc: stable@dpdk.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Always use the virtio variants which support the platform
memory ordering.
Fixes: 9230ab8d79 ("net/virtio: support platform memory ordering")
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Acked-by: Ilya Maximets <i.maximets@samsung.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch fixed below issues in the packed ring based control
vq support in virtio user:
1. The idx_hdr should be used_idx instead of the id in the desc;
2. We just need to write out a single used descriptor for each
descriptor list;
3. The avail/used bits should be initialized to 0;
Meanwhile, make the function name consistent with other parts.
Fixes: 48a4464029 ("net/virtio-user: support control VQ for packed")
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This patch mainly fixed below issues in the packed ring based
control vq support in virtio driver:
1. When parsing the used descriptors, we have to track the
number of descs that we need to skip;
2. vq->vq_free_cnt was decreased twice for a same desc;
Meanwhile, make the function name consistent with other parts.
Fixes: ec194c2f18 ("net/virtio: support packed queue in send command")
Fixes: a4270ea4ff ("net/virtio: check head desc with correct wrap counter")
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Function cmd_config_tx_metadata_specific_parsed() takes input value,
applies rte_cpu_to_be_32() on it, and stores it for Tx use.
Function cmd_show_tx_metadata_parsed() displays the stored value as is.
This patch modifies function cmd_show_tx_metadata_parsed(), to apply
rte_be_to_cpu_32() on the stored value before displaying it.
Fixes: c18feafa19 ("app/testpmd: support metadata as flow rule item")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Fixes: 6bc985e411 ("net/sfc: support TSO in EF10 Tx datapath")
Cc: stable@dpdk.org
Signed-off-by: Igor Romanov <igor.romanov@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Fix the issue that the link speed is unknown.
Fixes: 690175ee51 ("net/ice: support getting device information")
Fixes: cf911d90e3 ("net/ice: support link update")
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Enable the APIs for unicast and multicast promiscuous
mode setting.
Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
The original commit enabled the creation of flow rule with empty
ETH address and with specific VLAN.
It works fine on PF, but such flow rule is still rejected on VF due
to kernel limitation, which blocks MAC address of all zeroes.
This patch complements the original commit.
It sets byte 0 of MAC address to 0xFF.
Fixes: c0d2392631 ("net/mlx4: support flow w/o ETH spec and with VLAN")
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Shahaf Shuler <shahafs@mellanox.com>
Fix a possible out of bound access which may happen when handling
indirect descs in split ring.
Fixes: 1be4ebb1c4 ("vhost: support indirect descriptor in mergeable Rx")
Cc: stable@dpdk.org
Reported-by: Haiyue Wang <haiyue.wang@intel.com>
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
DCB is not necessary to be configured if no user priority of
queue region is requested.
Unexpected configuration of DCB will cause unexpected behavior, like
FDIR does not work on those queues, but could cause more issues.
Fixes: 7cbecc2f74 ("net/i40e: support queue region set and flush")
Cc: stable@dpdk.org
Signed-off-by: Wei Zhao <wei.zhao1@intel.com>
Tested-by: Yuan Peng <yuan.peng@intel.com>
Acked-by: Qi Zhang <qi.z.zhang@intel.com>
Dynamic memory mode allowes zero socket-mem because all the
required memory could be allocated on demand.
Fixes: 0a703f0f36 ("eal/linux: fix parsing zero socket memory and limits")
Cc: stable@dpdk.org
Reported-by: Shuai Zhu <shuaix.zhu@intel.com>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Modern memory mode allowes to not reserve any memory by the
'--socket-mem' option. i.e. it could be possible to specify
zero preallocated memory like '--socket-mem 0'.
Also, it should be possible to configure unlimited memory
allocations by '--socket-limit 0'.
Both cases are impossible now and blocks starting the DPDK
application:
./dpdk-app --socket-limit 0 <...>
EAL: invalid parameters for --socket-limit
EAL: Invalid 'command line' arguments.
Unable to initialize DPDK: Invalid argument
Fixes: 6b42f75632 ("eal: enable non-legacy memory mode")
Cc: stable@dpdk.org
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Build error seen with Linux kernel 5.0 and
when CONFIG_RTE_KNI_KMOD_ETHTOOL is enabled.
build error:
.../build/build/kernel/linux/kni/igb_main.c:2348:24:
error: initialization of
‘int (*)(struct net_device *, struct nlmsghdr *, u16,
struct netlink_ext_ack *)’
{aka ‘int (*)(struct net_device *, struct nlmsghdr *,
short unsigned int, struct netlink_ext_ack *)’}
from incompatible pointer type
‘int (*)(struct net_device *, struct nlmsghdr *, u16)’
{aka ‘int (*)(struct net_device *, struct nlmsghdr *,
short unsigned int)’}
[-Werror=incompatible-pointer-types]
.ndo_bridge_setlink = igb_ndo_bridge_setlink,
^~~~~~~~~~~~~~~~~~~~~~
.../build/build/kernel/linux/kni/igb_main.c:2348:24:
note: (near initialization for ‘igb_netdev_ops.ndo_bridge_setlink’)
igb_ndo_bridge_setlink() is changed in Linux kernel version 5.0
and now requires a new parameter, 'struct netlink_ext_ack *extack'.
Fixed by adding a new parameter with a kernel version check.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Build error seen with Linux kernel 5.0 and
when CONFIG_RTE_KNI_KMOD_ETHTOOL is enabled.
build error:
.../build/build/kernel/linux/kni/ixgbe_ethtool.c:1746:4:
error: too few arguments to function ‘dev_open’
dev_open(netdev);
^~~~~~~~
In file included from .../build/kernel/linux/kni/ixgbe_ethtool.c:18:
.../linux/linux/include/linux/netdevice.h:2620:5: note: declared here
int dev_open(struct net_device *dev, struct netlink_ext_ack *extack);
^~~~~~~~
.../build/build/kernel/linux/kni/igb_ethtool.c:1812:4:
error: too few arguments to function ‘dev_open’
dev_open(netdev);
^~~~~~~~
In file included from .../build/build/kernel/linux/kni/igb_ethtool.c:15:
.../linux/linux/include/linux/netdevice.h:2620:5: note: declared here
int dev_open(struct net_device *dev, struct netlink_ext_ack *extack);
^~~~~~~~
dev_open() is changed in Linux kernel version 5.0 and now requires
a new parameter, 'struct netlink_ext_ack *extack'.
Fixed by defining dev_open as macro when kernel version >= 5.0
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch fixes the missed digest and aad data physical
addresses filling to crypto operations in fips_validation
sample application.
Fixes: 41d561cbdd ("examples/fips_validation: add power on self test")
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
This patch fixes the inflexible compile issue AESNI-MB PMD.
Originally the compile of the PMD will fail if IPSec_MB is
not installed in default location.
Fixes: c68d7aa354 ("crypto/aesni_mb: use architecture independent macros")
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Tested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Tunnel type is an enum in PKT_TX_TUNNEL_MASK bits.
As the result, for example, IPIP or MPLSinUDP tunnel packets may be
incorrectly treated internally as VXLAN.
Fixes: b058d92ea9 ("gso: support VxLAN GSO")
Fixes: 70e737e448 ("gso: support GRE GSO")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Jiayu Hu <jiayu.hu@intel.com>
rte_errno should be set to positive value from errno.h plus
few RTE-specific values.
Fixes: 4fb7e803eb ("ethdev: add Tx preparation")
Fixes: 439a90b5f2 ("ethdev: reorder inline functions")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
It is only possible to know IOMMU type of a given VFIO container
by attempting to initialize it. Since secondary process never
attempts to set up VFIO container itself (because they're shared
between primary and secondary), it never knows which IOMMU type
the container is using, and never sets up the appropriate config
structures. This results in inability to perform DMA mappings in
secondary process.
Fix this by allowing secondary process to query IOMMU type of
primary's default container at device initialization.
Note that this fix is assuming we're only interested in default
container.
Bugzilla ID: 174
Fixes: 6bcb7c95fe ("vfio: share default container in multi-process")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Negative test case for passing invalid count size to
rte_metrics_update_values() will pass if count value
is any value greater than 1 because set size wrt to
key id 1 is 1 and passing 0 as no of values to be
updated will not hit the ERANGE check.
Fixes: cd38042429 ("test/metrics: add unit tests for metrics library")
Cc: stable@dpdk.org
Signed-off-by: Harman Kalra <hkalra@marvell.com>
Acked-by: Remy Horton <remy.horton@intel.com>
Add limitation that AAD buffer must be padded with zeroes
in GCM case.
Fixes: d4f44265a1 ("doc: add limitation of AAD size to QAT guide")
Cc: stable@dpdk.org
Signed-off-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Arek Kusztal <arkadiuszx.kusztal@intel.com>
With the addition of the patchset to allow auto-detection and use
of the intel_pstate kernel driver instead of the acpi-cpufreq kernel
driver, we need to reflect this in the documentation.
Now, instead of telling the user to disable intel_pstate in all cases,
it is now an optional step, should the user want to use acpi-cpufreq.
Signed-off-by: David Hunt <david.hunt@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
In the References section in the Power Management overview,
both links pointed to the same l3fwd-power app. Fix the links
so that one points to l3fwd-power, and the other points to
the vm_power_manager sample app.
Signed-off-by: David Hunt <david.hunt@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
Update recommended BIOS settings for performance to be more appropriate
for a wider variety of applications, and allow users to select the most
appropriate settings for their use case, e.g. some users may wish to
have Turbo Boost enabled.
Signed-off-by: David Hunt <david.hunt@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
No need to keep those file listings, they are very likely to become
outdated.
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Marko Kovacevic <marko.kovacevic@intel.com>
Update the Linux user guide to restrict the supported kernels
to reasonnably recent enough versions.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Categorize testcases into suites:
- test cases that runs quickly
- performance test cases
- test cases that depend on specific drivers
- logging or dump related test cases
Signed-off-by: Hari Kumar Vemula <hari.kumarx.vemula@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Incorrect placement of prefetch in Tx path is causing
a performance drop of around ~2% on AMD platform.
Fixes: 2ea6f76aff ("qede: add core driver")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shshaikh@marvell.com>
Acked-by: Rasesh Mody <rmody@marvell.com>
Allocating replacement buffer per received packet is expensive.
Instead, process received packets first and allocate
replacement buffers in bulk later.
This improves performance by ~25% in terms of PPS on AMD
platforms.
Fixes: 2ea6f76aff ("qede: add core driver")
Cc: stable@dpdk.org
Signed-off-by: Shahed Shaikh <shshaikh@marvell.com>
Acked-by: Rasesh Mody <rmody@marvell.com>
When ifpga_rawdev_create() allocate memory for a new rawdev,
the original code allocate redundant memory for adapter,
which is a member of the rawdev. What is actually necessary is
the adapter to be initialized, not memory allocated.
Fixes: ef1e8ede3d ("raw/ifpga: add Intel FPGA bus rawdev driver")
Cc: stable@dpdk.org
Signed-off-by: Andy Pei <andy.pei@intel.com>
Acked-by: Tianfei Zhang <tianfei.zhang@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
In the original code, when an AFU device probe a driver,
if the first driver in the driver list does not support
this device or some error happens, bus probe returns an error.
With this patch, a device will try to match driver in the driver
list one by one until an appropriate driver is found.
If some error happens, the error is returned. If the current driver
does not support the device, just try the next driver in the list.
If all the drivers in the list are tried and no driver matches,
0 is returned. Otherwise, the error code is returned.
Fixes: 05fa3d4a65 ("bus/ifpga: add Intel FPGA bus library")
Cc: stable@dpdk.org
Signed-off-by: Andy Pei <andy.pei@intel.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
Original driver requires users to input "afu_bts" args.
But driver can also work without an explicit "afu_bts" args.
Fixes: 05fa3d4a65 ("bus/ifpga: add Intel FPGA bus library")
Cc: stable@dpdk.org
Signed-off-by: Andy Pei <andy.pei@intel.com>
Acked-by: Rosen Xu <rosen.xu@intel.com>
This patch removes an unnecessary flag which was breaking the
build with older versions of the ISA-L library (v2.23 and older)
and replace with a more appropriate flag which is present
in older versions of library.
Fixes: bd03d3f1e4 ("compress/isal: enable checksum support")
Signed-off-by: Lee Daly <lee.daly@intel.com>
Tested-by: Harry van Haaren <harry.van.haaren@intel.com>
In rte_efd_create() allocated memory for tail queue entry but
not freed.
Added freeing the tail queue entry.
Fixes: 56b6ef874f ("efd: new Elastic Flow Distributor library")
Cc: stable@dpdk.org
Signed-off-by: Hari Kumar Vemula <hari.kumarx.vemula@intel.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Acked-by: Reshma Pattan <reshma.pattan@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Currently this api doesn't report error if name is
truncated and so user is not sure about uniqueness
of name. This change reports error to help user.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
This patch adds a missing topology "loop" mode for port-topology
option in testpmd guide. The testpmd "loop" option
(PORT_TOPOLOGY_LOOP) was added to testpmd parameter
port-topology back in 2014, but it seems that it was never added
to the testpmd guide. This patch adds it in the testpmd guide.
Fixes: 3e2006d618 ("app/testpmd: add loopback topology")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
This patch fixes an error in ifc NIC document; a previous patch
changed the semantics to use CONFIG_RTE_LIBRTE_IFC_PMD
instead of CONFIG_RTE_LIBRTE_IFCVF_VDPA_PMD,
but the ifc NIC doc file remained with the old name.
Fixes: 4b614e9504 ("net/ifc: make driver name consistent")
Cc: stable@dpdk.org
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Errors, missed and pauses should not be included in delivered
good Rx packets reported by ipackets.
Fixes: 1caab2f1e6 ("net/sfc: add basic statistics")
Cc: stable@dpdk.org
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
This patch fixes the incorrect session private mempool passing
to cryptodev.
Fixes: ac5e42daca ("vhost/crypto: use separate session mempools")
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Add the missing failure handling for path allocation,
as realloc() may fail.
Fixes: ad0eef4d22 ("examples/vhost: support multiple socket files")
Cc: stable@dpdk.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
vhost_user_host_notifier_ctrl is not existed anymore, its statement in
header file should be removed accordingly.
Fixes: 43f34e3566 ("vhost: provide helper for host notifier ctrl")
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Reviewed-by: Tiwei Bie <tiwei.bie@intel.com>
As qemu will only send VHOST_USER_SET_VRING_ENABLE message for guest
enabled vrings (only first queue pair will be enabled at initialized
stage), this will cause trouble for multiqueue case, vDPA's dev_conf
callback will get no chance be invoked. Decouple the dev_conf callback from
VHOST_USER_SET_VRING_ENABLE solves this issue.
Signed-off-by: Xiaolong Ye <xiaolong.ye@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Free flow profile entries when free HW tables.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Signed-off-by: Paul M Stillwell Jr <paul.m.stillwell.jr@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
